About Dissertation Research:
Essays on Employee Management in Service Operations
The service sector is the largest and the fastest-growing segment in the economies of developed nations, including the USA (Chase and Apte 2007). Thus, one cannot get enough by underscoring the immense significance of service operations in shaping the economic landscape of modern nations and the potential repercussions of not recognizing its importance.
However, as anticipated, there are notable challenges tied to enhancing profitability in services, particularly without increasing labor hours. An effective way to improve service operations involves empowering employees with skills, enabling them to serve customers more effectively. After all, highly satisfied customers are key drivers of growth and profitability in service-oriented organizations. The service profit chain, as proposed by Heskett et al. (1994), establishes a connection between a service company's profitability and growth with the satisfaction and loyalty of its employees. According to this model, customer loyalty drives profitability and growth, and this loyalty stems from the perceived value of the service. A critical component for offering valuable service is having well supported, and efficient employees. Thus, for a service organization to thrive, it needs to focus on employee-related factors that effects the entire operation and has significant impact on the organization’s profitability.
With this motivation and perspective, this dissertation takes an employee-oriented approach to the within-firm OM decisions and investigates the effects of interventions focusing on employees on the process outcomes. Our approach is to understand challenges beyond successful implementation of the tools that’s proven to be challenging yet rewarding once done. In general, our approach is to better understand and empower employees and support them with feedback, knowledge and technology to increase their engagement, satisfaction and loyalty through which service organizations could deliver operational outcomes via expanding satisfaction and loyalty of customers.
For this dissertation work, I study three different employee management tools that have a significant effect on operational outcomes such as task performance and customer satisfaction. These tools are rank-based relative performance feedback (RPF), knowledge transfer via best practices adoption (BPA), and artificial intelligence (AI) based algorithmic feedback and real-time coaching. The common divider among all these tools is that they pose significant challenges beyond successful implementation and get resistance from employees in the form of negative emotional responses, rejection, resistance for acceptance or disengagement. We seek to understand the incremental effect of providing these tools on employee performance via changes in employee behavior when they are used in combination or separately. Ultimately our goal is to discover ways in which these management tools affect employee behavior, stimulate acceptance and engagement, which in turn affects customer experience and operational outcomes, thereby fostering more efficient and productive systems.
The Demotivating Effects of Relative Performance Feedback on Middle-Ranked Workers’ Performance
joint work with Prof. Anita Carson, submitted to the Management Science
Click here to see the manuscript in SSRN
This work examines how various forms of RPF influence worker productivity and what underlies the change. We conduct a series of experiments to unravel the effects of three different types of RPF on middle-ranked workers’ output on a skill-based task. In our study, participants do not receive any information to help them improve at the task and there are no financial incentives. Intriguingly, our findings suggest that the provision of feedback, regardless of its nature, tends to decrease worker performance in complex tasks. We find that receiving any type of feedback reduces performance compared to no feedback. We conduct mediation analysis and show that receiving feedback changes employees’ feelings associated with general performance, which explains the performance reduction. Aligned with theory, delivering feedback increases the focal employee’s social comparison involvement (SCI), which measures the focal individual’s tendency for comparing themselves to others while performing the task, and their shame. Our results imply that making individuals care more about social comparisons and creating feelings of shame decreases performance on skill-based tasks for middle-ranked workers in the absence of financial incentives and information on how to improve. An implication of our study is that any form of rank-based performance feedback should be implemented with caution as it may harm the performance of the majority of workers.
Keywords : worker performance, relative performance feedback, online experiments, behavioral operations, rank feedback
Effects of Relative Performance Feedback on Adoption of Best Practices and Worker Performance
joint work with Prof. Anita Carson, in preparation
The second chapter, co-authored with Anita Carson, concentrates on enhancing performance through fostering internal knowledge transfer and promoting the adoption of best practices. Our earlier studies demonstrated that the mere provision of performance feedback fails to boost performance without equipping employees with tools to hone their skills or approaches. Through a series of experiments, we assess the effects of providing performance feedback in conjunction with best practices on knowledge-seeking behavior, best practice adoptions, and operational performance. We further investigate whether it is feasible to transform the negative implications of feedback into positive outcomes, using knowledge stemming from superior-performing peers that provide actionable insights into improving task performance rather than merely exerting greater effort. Our study poses an exciting finding by showing that RPF's previously documented negative effect on middle-ranked workers could be mitigated, and performance improvement could be attained when combined with best practices. Our research provides implications for decision-makers to utilize more effective strategies for motivating their employees with rank-based feedback and advert the previously documented negative consequences on middle-ranked workers.
Artificial Intelligence in Customer Service Operations
joint work with Asst. Prof. Michelle Shell, in preperation
The concluding chapter, co-authored with Michelle Kinch, focuses on the effect of using algorithmic feedback and coaching as management tool in service operations within call centers, which has been a major channel for customer-firm interactions (Aksin et al. 2007, Tezcan and Behzad 2012). New developments, and recent advancements in information technology offer opportunities to redesign and improve service-delivery operations within these centers (Wang et al. 2022). Among these, companies are deploying Artificial Intelligence applications into service settings in a variety of ways, from automating agent tasks to replacing human servers altogether. AI-supported customer service plays a pivotal role in propelling the expansion of the global AI market, anticipated to surge from $6.8 billion in 2021 to $18.4 billion by 2026 (Markets and Markets 2021, Wang et al. 2022). Additionally, the AI market stood at $959 million in 2020 and is forecasted to escalate to $9.95 billion by 2030 (Reports 2022). Given this substantial economic activity and growth trajectory, ensuring efficient service in establishing these systems is of utmost importance. This study examines how artificial intelligence-based feedback (AI) impacts customer service agent employee productivity as measured by three key performance indicators: call-handle time, customer satisfaction, and call service quality. Our field partner, a North American outsourced call center deployed the AI software to monitor calls during a bill collection campaign and provide visible cues to remind agents of their service script requirements. In this way, the AI acts as a real-time supervisor, assessing agent performance and offering real-time feedback during and after the call. This contrasts with the traditional approach where supervisors periodically listen to call samples, evaluate performance and quality, and use otherwise productive agent capacity for coaching and feedback sessions. Using international call center data, we provide evidence that agents with access to the AI feedback are indeed more likely to comply with scripts and in so doing, deliver increased operational efficiency with lower call handle time. Moreover, calls conducted with AI feedback show an increase in two service quality metrics not commonly associated with technology-assisted communication: respect and rapport. This research presents evidence of successful human-machine collaboration without theorized algorithmic aversion or decline in customer satisfaction - effectively challenging the conventional service-efficiency trade-off and offering practical insights for companies seeking to augment service encounters with artificial intelligence technologies.
Keywords; empirical operations, artificial intelligence, service operations, call centers, customer service
Our findings suggest that a frequently used management tool of providing rank-based feedback can hinder the performance of mid-tier workers, especially when lacking financial incentives and guidance for improvement, as, by itself, rank-based feedback emphasizes social comparisons and induces shame on average employees, which constitutes most of the workforce in an organization. Thus, decision-makers shall tread carefully with rank-based feedback, as it could adversely affect most of their employees. However, this negative impact can be mitigated and even reversed when feedback is complemented by another management tool to provide guidance for improvement on tasks; internal knowledge transfer via the adoption of best practices. This way, with the interdependent provision of these two tools, challenges ahead of successful implementation can be mitigated, compared to offering each in isolation. Finally, our findings of implementing algorithmic feedback as a management tool showcase a successful human-machine collaboration through improved employee performance in operations without signs of diminished customer experience, offering managers practice implementation on effectively breaking the conventional service-efficiency trade-off from an employee's perspective. In summary, through three studies, we offer theoretical and practical implications about the use and challenges associated with various management tools and provide ways to improve employee behavior to stimulate engagement and foster more efficient and productive systems.